Neighborhood Sorting Dynamics in Real Estate: Evidence from the Virginia Sex Offender Registry

Abstract

Given the potential risk of recidivism, recent research has found that registered sex offenders impose external costs that are capitalized into the value (lower prices) and liquidity (longer time on market) of nearby residential real estate. Using more than a decade of real estate listings data and a unique data set of registered sex offenders (RSOs) in Virginia, we extend this literature by investigating how real estate markets respond to clustering of this particular group and whether there is evidence of market prices facilitating clustering phenomena. Employing a difference-in-difference approach within a hedonic empirical framework, the results show that the housing market has a nonlinear response to clusters, as it considerably discounts prices of homes located near a cluster (4+) of RSOs, providing a market incentive for further sorting/clustering. Our empirical findings and supplemental geospatial analysis are consistent with a sorting/tipping model in which individuals choose residential locations according to preferences and price incentives. Insofar as clustering of negative externalities is efficient, we further discuss the market efficiency implications in this context.

Keywords: externalities, residential real estate, neighborhood tipping, sorting, sex offenders, spatial cluster

JEL Classification: R12, H23, R23, D62